On Wednesday, January 21, 2015 05:05:36 PM Alan Stern wrote: > On Wed, 21 Jan 2015, Rafael J. Wysocki wrote: > > > > Well, assuming that the HCD driver doesn't support wakeup signaling, > > > it would be good to have a way to indicate the lack of support for > > > autosuspend to the USB core so as to avoid the user-visible breakage. > > > > > > Do we have any means for that? > > > > For example, would it help to add > > > > device_set_wakeup_capable(hcd->self.controller, 0); > > > > somewhere at the beginning of isp1760_hc_setup()? > > Yes, we could use that instead of adding a new flag. > > > For example, would it help to add > > > > device_set_wakeup_capable(hcd->self.controller, 0); > > > > somewhere at the beginning of isp1760_hc_setup()? > > At the moment it wouldn't help. The problem isn't that we want to > avoid suspending the controller or its root hub; the problem is that we > want to avoid suspending a device plugged into the root hub if that > device needs to be enabled for wakeup. This is because when the device > sends its wakeup signal, the controller won't receive it. I understand that. Well, OK, so there is the case when the controller is not suspended, but it doesn't generate interrupts for wakeup signals from devices, which generally is different from the case when the controller itself is suspended too that requires it to be wakeup-capable for things to work. An additional flag may still be better then. Rafael -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
